An Introduction to the Physical Basis for Deriving River Information by Optical Remote Sensing - Fluvial Remote Sensing for Science and Management - page 42

3

An Introduction to the Physical
Basis for Deriving River
Information by Optical Remote
Sensing

Carl J. Legleiter 1 and Mark A. Fonstad 2
1 Department of Geography, University of Wyoming, Laramie, WY, USA
2 Department of Geography, University of Oregon, Eugene, OR, USA

can be derived from river images, as well as the reliability
of that information. The goal of this chapter is to provide
an overview of the physical processes that underlie the
application of remote sensing to rivers. Our objective is to
equip prospective users of image data with knowledge of
these processes and thus prepare them to make informed,
rational assessments of what can and cannot be achieved
via remote sensing. Such knowledge, we believe, is the key
to realising this technology's widely recognised but largely
untapped potential to contribute to river management.
Remote sensing of rivers is a novel field of inquiry
that remains in an early stage of development relative to
remote mapping of coastal environments, where sophis-
ticated, physics-based approaches are now favored (Lee
et al., 2001; Mobley et al., 2005; Lesser and Mobley,
2007). Stream studies, in contrast, tend to rely upon
more empirical techniques. Typically, image pixel values
are related to ground-based measurements of channel
attributes - depth, suspended sediment concentration,
bed material grain size, etc. - to establish statistical cor-
relations. The resulting regression equations are then
applied to pixels throughout the image to map the quan-
tities of interest. This approach has been implemented
successfully (e.g., Winterbottom and Gilvear, 1997; Lejot
et al., 2007, compilation of results in Chapter 2) but is

3.1 Introduction

Remote sensing affords considerable potential to not only
foster significant advances in our scientific understand-
ing of river systems but also facilitate monitoring and
management of these crucial resources. The chapters in
this topic illustrate the range of ways in which this tech-
nology has been applied to diverse fluvial settings. This
spectrum of applications can be expected to expand as
new sensors, new methods, and new uses of remotely
sensed data are developed in the coming years (Marcus
and Fonstad, 2008). While the successes documented in
this volume (Marcus et al., Chapter 2) and in the pri-
mary literature (reviewed by Feurer et al., 2008) justify
a degree of optimism regarding these possibilities, a cer-
tain amount of caution is required as well. The key to
efficient, effective use of remote sensing techniques in
riverine environments is an appreciation of the physical
processes that govern the interaction of light and water
in stream channels. These processes, collectively referred
to as radiative transfer, enable the attributes of inter-
est to managers to be inferred from various kinds of
image data. At the same time, the mechanics of radiative
transfer, along with certain sensor characteristics, impose
fundamental limitations as to what kind of information